In this case study, we will demonstrate a common application of multivariate analysis: search for a pattern in a set of vegetation records with available measurements of environmental conditions. The primary data (courtesy of Michal Hájek – see Hájek et al. 2002) comprise classical vegetation relevés (records of all vascular plants and bryophytes, with estimates of their cover using the Braun-Blanquet scale) in spring fen meadows in the mountain ranges of the westernmost Carpathians. The data represent an ordinal transformation of the Braun-Blanquet scale, i.e. the r, +, and 1 to 5 values of the Braun-Blanquet scale are replaced by numeric values 1 to 7. The data presented here were extracted from a database using the TURBOVEG program (Hennekens & Schaminee 2001).

The relevés are complemented with environmental data – chemical analyses of the spring water, soil organic carbon content, and slope of the locality. The ion concentrations represent their molarity: remember that predictors are standardised in all Canoco procedures and, consequently, the choice of units does not play any role, as long as the various units are linearly related. The distributions of ion concentrations are highly skewed, i.e. there are usually few highly positively outlying values, which would have a strong effect on the results. Moreover, we can expect that the vegetation composition will be better related to the logarithm of concentrations (i.e. the size of the response will be similar when we double the concentration rather than when we add a constant amount). Consequently, we have first log-transformed ion concentrations, but not the organic carbon (Corg), pH, conductivity, and slope. The transformation was done directly in the Excel file before import. We will use the log-transformed values in the analyses. The aim of this case study is to describe basic vegetation patterns and their relationship with available environmental data, mainly with the chemical properties of the spring water.